Connecting Claude to a WordPress site over the Model Context Protocol turns your posts and comments into something it can read. Here is the setup, the gap-analysis prompt I use, and how I score the ideas that come back.
Vector databases have become the default answer for AI agent memory, and for a project-scoped agent they are usually the wrong one. Plain Markdown files plus a local SQLite index give you grep, git diff and no server to babysit.
LLM agent memory architecture is what decides whether an agent holds up over a long task. This covers the four temporal scopes, the manage step almost everyone skips in the write-manage-read loop, and how summarization drift creeps in.
Six things I ran into building a GPT-2 clone from scratch in PyTorch: the scaling flaw RsLoRA fixes, why RoPE replaced additive positional embeddings, why weight tying stopped paying off, and what KV-caching costs you in VRAM.
Notes from WordCamp Asia 2026 in Mumbai: plugins and themes landed on every paid WordPress.com plan, the speed build put AI against raw FSE work, and Nirav Mehta’s session covered where LLMs still break. Here is what held up as useful for real builds.
Claude Code is not only for refactoring PHP. Three things I run through it that never touch a codebase: LaTeX decks kept in Git, cross-app search over Slack and Notion, and prospect research.
LLMs forget your stack the moment you close the tab. Here is what actually sticks: a project rules file in the repo, global rules for your own preferences, and MCP when the assistant needs data from outside the codebase.
One LLM cannot draft emails, read documents and optimize a sales pipeline well. A look at splitting the work across reasoning, orchestration and execution layers, with TD learning doing the sequential decisions.
Voice picking used to mean vendor-locked hardware. ElevenLabs voice AI does the same job off an ordinary smartphone, and this covers the WordPress side of it: a cached text-to-speech wrapper, plus the latency and noise problems that sink most builds.
Vector search on its own leaves an agent repeating things that stopped being true months ago. Here is the memory lifecycle I build instead: decay scores, contradiction detection, confidence values, compression, and a background worker that does the cleanup.